Agent READMEs: An Empirical Study of Context Files for Agentic Coding
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914160432906240 |
|---|---|
| author | Chatlatanagulchai, Worawalan Li, Hao Kashiwa, Yutaro Reid, Brittany Thonglek, Kundjanasith Leelaprute, Pattara Rungsawang, Arnon Manaskasemsak, Bundit Adams, Bram Hassan, Ahmed E. Iida, Hajimu |
| author_facet | Chatlatanagulchai, Worawalan Li, Hao Kashiwa, Yutaro Reid, Brittany Thonglek, Kundjanasith Leelaprute, Pattara Rungsawang, Arnon Manaskasemsak, Bundit Adams, Bram Hassan, Ahmed E. Iida, Hajimu |
| contents | Agentic coding tools receive goals written in natural language as input, break them down into specific tasks, and write or execute the actual code with minimal human intervention. Central to this process are agent context files ("READMEs for agents") that provide persistent, project-level instructions. In this paper, we conduct the first large-scale empirical study of 2,303 agent context files from 1,925 repositories to characterize their structure, maintenance, and content. We find that these files are not static documentation but complex, difficult-to-read artifacts that evolve like configuration code, maintained through frequent, small additions. Our content analysis of 16 instruction types shows that developers prioritize functional context, such as build and run commands (62.3%), implementation details (69.9%), and architecture (67.7%). We also identify a significant gap: non-functional requirements like security (14.5%) and performance (14.5%) are rarely specified. These findings indicate that while developers use context files to make agents functional, they provide few guardrails to ensure that agent-written code is secure or performant, highlighting the need for improved tooling and practices. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_12884 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Agent READMEs: An Empirical Study of Context Files for Agentic Coding Chatlatanagulchai, Worawalan Li, Hao Kashiwa, Yutaro Reid, Brittany Thonglek, Kundjanasith Leelaprute, Pattara Rungsawang, Arnon Manaskasemsak, Bundit Adams, Bram Hassan, Ahmed E. Iida, Hajimu Software Engineering Agentic coding tools receive goals written in natural language as input, break them down into specific tasks, and write or execute the actual code with minimal human intervention. Central to this process are agent context files ("READMEs for agents") that provide persistent, project-level instructions. In this paper, we conduct the first large-scale empirical study of 2,303 agent context files from 1,925 repositories to characterize their structure, maintenance, and content. We find that these files are not static documentation but complex, difficult-to-read artifacts that evolve like configuration code, maintained through frequent, small additions. Our content analysis of 16 instruction types shows that developers prioritize functional context, such as build and run commands (62.3%), implementation details (69.9%), and architecture (67.7%). We also identify a significant gap: non-functional requirements like security (14.5%) and performance (14.5%) are rarely specified. These findings indicate that while developers use context files to make agents functional, they provide few guardrails to ensure that agent-written code is secure or performant, highlighting the need for improved tooling and practices. |
| title | Agent READMEs: An Empirical Study of Context Files for Agentic Coding |
| topic | Software Engineering |
| url | https://arxiv.org/abs/2511.12884 |